On Tue, Dec 29, 2015 at 1:12 PM, Ganesh Ajjanagadde wrote:
> On Tue, Dec 29, 2015 at 11:29 AM, Clément Bœsch wrote:
>> On Tue, Dec 29, 2015 at 09:28:34AM -0800, Ganesh Ajjanagadde wrote:
>>> The table is highly structured, so pow (or exp2 for that matter) can
>>> entirely
>>> be avoided, yieldin
On Tue, Dec 29, 2015 at 11:29 AM, Clément Bœsch wrote:
> On Tue, Dec 29, 2015 at 09:28:34AM -0800, Ganesh Ajjanagadde wrote:
>> The table is highly structured, so pow (or exp2 for that matter) can entirely
>> be avoided, yielding a ~ 40x speedup with no loss of accuracy.
>>
>> sample benchmark (Ha
On Tue, Dec 29, 2015 at 09:28:34AM -0800, Ganesh Ajjanagadde wrote:
> The table is highly structured, so pow (or exp2 for that matter) can entirely
> be avoided, yielding a ~ 40x speedup with no loss of accuracy.
>
> sample benchmark (Haswell, GNU/Linux):
> new:
> 4449 decicycles in init_pow2table
The table is highly structured, so pow (or exp2 for that matter) can entirely
be avoided, yielding a ~ 40x speedup with no loss of accuracy.
sample benchmark (Haswell, GNU/Linux):
new:
4449 decicycles in init_pow2table(loop 1000), 254 runs, 2 skips
4411 decicycles in init_pow2table(loop 1
